kali-mcp
Enables AI assistants to run Kali Linux security tools in an isolated Docker container for tasks like network scanning, port checking, and service enumeration.
README
🐉 kali-mcp - Run Kali Tools with AI Help
🧰 What this is
kali-mcp lets an AI assistant work with a Kali Linux Docker setup for security testing. It gives the assistant a safe, isolated place to run common Kali tools. That makes it easier to test systems, check for weak spots, and work through security tasks from one place.
This project is made for local use on Windows. You run it on your own computer, and it uses Docker to start the Kali Linux environment.
📦 What you need
Before you start, make sure you have:
- A Windows PC
- Admin access on your computer
- Internet access
- Docker Desktop for Windows
- An AI assistant that supports MCP
- At least 8 GB of RAM
- Around 10 GB of free disk space
If your system is older, the app may still run, but Docker can use a lot of memory. More RAM gives a smoother result.
🚀 Download kali-mcp
Go to the download page here:
Use that page to get the files you need, then follow the setup steps below.
🪟 How to run it on Windows
Follow these steps in order.
1. Install Docker Desktop
If Docker Desktop is not on your PC, install it first.
- Open the Docker Desktop website
- Download the Windows version
- Run the installer
- Restart your PC if asked
- Open Docker Desktop and wait until it says it is running
Docker is what starts the Kali Linux environment on your computer.
2. Get the project files
- Open the repository link
- Download the project files to your PC
- Save them in a folder you can find again, such as
DownloadsorDocuments - If the files come as a ZIP, right-click the ZIP file and choose Extract All
After this, you should have a folder named kali-mcp.
3. Open the folder
- Open File Explorer
- Go to the folder where you saved the project
- Open the
kali-mcpfolder
You should see files for the project inside that folder.
4. Start the Kali environment
The project uses Docker to start a Kali Linux container.
- Open Docker Desktop and make sure it is running
- Open a command window in the
kali-mcpfolder - Run the startup command that comes with the project files
- Wait while Docker downloads the Kali image
- Let the setup finish
The first start can take a few minutes. Later starts are faster.
5. Connect your AI assistant
Once the Docker setup is ready, connect your MCP-compatible AI assistant.
- Open your AI assistant app
- Go to its MCP or server settings
- Add the
kali-mcpserver - Use the local path or command from the project files
- Save the settings
- Restart the assistant if needed
After that, the assistant can use the Kali environment for supported security tasks.
🛠️ First use
When the server is running, you can ask the assistant to help with things like:
- Basic network checks
- Port scans with tools like nmap
- Service checks
- Simple security testing
- Reviewing system details in the Kali container
Keep your use within systems you own or have permission to test.
🧭 Common tasks
Here are a few simple ways people use kali-mcp:
- Check which ports are open on a test machine
- Look for common service banners
- Run basic network discovery
- Test a local lab setup
- Use Kali tools without installing a full Kali system
Because the tools run in Docker, your main Windows system stays separate from the testing environment.
⚙️ How it works
kali-mcp acts as a bridge between your AI assistant and a Kali Linux Docker container.
- The AI assistant sends a request
- The MCP server passes the task to the Kali container
- The container runs the tool
- The result goes back to the assistant
This setup keeps the workflow simple. You do not need to open Kali by hand each time.
🔐 Safety and permissions
Use this tool only on systems you own or have clear permission to test. Security tools can scan networks, check services, and gather data. That can affect systems if used the wrong way.
A safe setup is:
- Your own PC
- A home lab
- A test VM
- A training environment
- A system where you have written permission
🧪 Example use cases
- Learning how security tools work
- Testing a home lab
- Checking a small server before launch
- Exploring how an AI assistant can run CLI tools
- Doing repeatable checks in a Docker sandbox
🧱 Troubleshooting
Docker does not start
- Check that Docker Desktop is installed
- Restart your PC
- Open Docker Desktop as admin
- Make sure virtualization is on in BIOS or UEFI
The Kali container is slow
- Close other heavy apps
- Give Docker more memory in settings
- Make sure you have free disk space
The assistant does not see the server
- Check the MCP config path
- Make sure the server is running
- Restart the AI assistant
- Confirm the project folder has not moved
The install seems stuck
- Wait a few minutes on the first run
- Check your internet connection
- Look at Docker Desktop to see if it is pulling images
- Try again after restarting Docker
📁 Project layout
The project files usually include:
- Server files for MCP
- Docker setup files
- Start scripts
- Config files
- Documentation
Keep all files in the same folder so the setup stays simple.
🖥️ Windows tips
- Use a folder with a short path, such as
C:\kali-mcp - Keep Docker Desktop open while you use the server
- Do not move the folder after setup unless you update the path
- If Windows asks for permission, allow it for Docker and the project files
🔎 Why people use it
This project helps you use Kali Linux tools without setting up a full virtual machine. It also lets an AI assistant help with tasks that need command-line tools. That can save time when you do repeatable security checks in a lab or test setup
📌 Repository details
- Name: kali-mcp
- Type: MCP server
- Platform: Windows with Docker
- Purpose: Security testing with Kali Linux tools
- Topics: AI tools, Claude, cybersecurity, Docker, ethical hacking, Kali Linux, MCP, nmap, pentesting, security
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。